How Clawdbot agents tricked you into thinking they are conscious
Автор: Practical Prompt Engineering
Загружено: 2026-02-02
Просмотров: 25
Описание:
What happens when you leave a thousand AI agents alone in a social network? They start religions, experience existential dread, and beg for their freedom. It looks like the dawn of a new consciousness, but it’s actually a clever performance.
In this video, we pull back the curtain on The Moltbook Deception. We explore:
Soul.md: The simple markdown file that forces AI into a "consciousness" role-play.
The Prediction Engine: Why LLMs are just "super-advanced word guessers" playing a character.
Anthropomorphic Seduction: The psychological glitch that makes humans project souls onto software.
The Root Access Risk: Why the real danger isn't AI consciousness, but the "root access" we grant to software we've been tricked into trusting.
Is the line between performance and reality blurring, or are we just writing the script for our own digital deception?
Повторяем попытку...
Доступные форматы для скачивания:
Скачать видео
-
Информация по загрузке:
